© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 31052
    Prev
    Next
    INDEX
    Explanations

    terms related to obstructions, obstacles, or hindrances

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    274
    +0.15
    0.9%
    111
    +0.13
    0.7%
    323
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    432
    +0.15
    0.02
    198
    +0.13
    0.01
    274
    +0.13
    0.01
    Negative Logits
    poons
    -1.70
    hip
    -1.69
    ľĵ
    -1.56
    azure
    -1.53
    fork
    -1.52
    mith
    -1.51
    hips
    -1.51
    Ķ
    -1.51
    hire
    -1.47
    axel
    -1.41
    POSITIVE LOGITS
    vi
    1.69
    nox
    1.68
    onents
    1.54
    enso
    1.50
     violated
    1.48
    uchi
    1.47
    amacare
    1.46
    estly
    1.44
    liography
    1.42
     infring
    1.40
    Activations Density 1.576%

    No Known Activations